Cathode bias setting in inefficient amps (2x6L6 > 20 watts)

[Pedro Vecino]

A question about adjusting cathode biased amps:
I´ve an old P.A. 20 watts amp that uses 360 volts and 2x6L6 cathode biased with a common 270 Ohm resistor + 22uF paralleled.
With a new pair of 6L6GC Sylvania (original model that had that amp but new), the current through each plate (directly measured) is 45mA.
The question is: when adjusting an amplifier like this, is considered maximum dissipation 25/30 watts per 6L6?, or should be considered the limit set (in this case 10/12 watts per 6L6)? Or another set by this low HT?
The output transformer has a size corresponding with 20 watts (deluxe reverb or similar) and with the original bias setting basically sounds good but tends to distortion prematurely.
If the amp had been 2x6V6 would not doubt about it, but having 2x6L6 ... That is the question. I'm not familiarized with this type of amplifiers.
Thank you very much!

[ThermionicScott]

The original circuit was surely designed around 19W 6L6's. I certainly wouldn't increase the idle current just because you have 6L6GC's in there now -- it would just reduce the clean headroom and make more heat. Since the tubes are running pretty cool now, you could try a more efficient rectifier tube for more headroom.

- Scott

[Pedro Vecino]

Thanks!. Very grateful.
The amp is a Bouyer ST20 (made in France, in 1960 approx.). Sockets are tagged "6L6" with little metallic squares attached. In 1978 I had another Bouyer model (ST75) that incorporated 4x6L6GC Sylvania, so I thought it might be the original tube model of these amps...
It uses a GZ34 (5AR4), therefore can not get more voltage with tube rectifiers.
The power it offers is similar to that of a Fender Deluxe and distortion levels, similar (a little more clear on this amp). Is recovered in the basics (filter caps, tubes, cathode bypass capacitors, electric protection...)
Could be a good idea to implant a selectable fixed bias adjustment to try to raise the clean threshold?
This is the amp: ST20 (ST 20) Radio Bouyer, Paul;
Here is the schematic: http://www.radiomuseum.org/forumdata...0st%2020kl.JPG
I suppose the system is the same: consider 20 watts..
Best regards

[Enzo]

If you want to use it for guitar, then wonderful. But it was designed as a PA amp, so they were not looking to bias the thing any hotter, as overdrive and breakup were not desired outcomes.

[ThermionicScott]

That's a neat little unit. I love the skull-and-crossbones on the meter.

I was a little off earlier -- it appears to have been designed with 6L6GC's in mind, but conservatively enough that you could plug any other 6L6 in there. Very interesting of them to use a triode-wired EF86 as the paraphase!

Converting to fixed-bias would increase the headroom a little, but you'd need a handful of other mods in concert:

1) modify the phase inverter to keep the negative bias voltage from screwing up its operation
2) bigger OT would be necessary
3) replace the 3.9K dropping resistor with a 1K or a choke to squeeze out a little more power

- Scott

[Pedro Vecino]

Thanks again!
Yes, Enzo. Amp is for guitar. It has already adjusted the sensitivity on ECC83 channel for guitar. And the filters for a good tone shaping.
Regarding the fixed bias ... had not noticed the type of driver (about the difficulty of implementing with it). Better to let original. The amp has a very nice old flavor with a particular compression.
Yes. I noticed the high resistor value for screen grids. Once I redo inputs and speaker outputs (this amp is a hazard) I'll try with different voltages on the screen grids, although I imagine that the limit will be close.
Something that caught my attention (I should have started looking at the spec sheets!) is this:



Relationship between exploited power and the OT impedance. When done other things, I will check the transformer impedance ratio.
It could be that selecting a different speaker impedance another performance was obtained. Later...
Regards